Add several lints into unused lint group

[petrochenkov]

Also a couple of obsolete (not reported) lints are removed.

r? @oli-obk

[petrochenkov]

@oli-obk
I initially wanted to move const_err into the future_incompatible group based on this comment, but it looks like this comment was actually outdated and this lint is not going to be an error?

[oli-obk]

Yea. These are essentially bugs

[petrochenkov]

I've also audited the remaining ungrouped lints.
In principle new lint groups can be created for them, but it looks like fine-grained lint grouping didn't find its use even in clippy, so I didn't do anything. cc clippy people just in case @mcarton @llogiq

Unused++.
These can also be quite reasonably added into the unused group, but less obviously than those I added into it in this PR.

STABLE_FEATURES
RENAMED_AND_REMOVED_LINTS
UNKNOWN_LINTS
UNUSED_COMPARISONS

Bad style++.
Probably can be added into the bad_style group, but it currently consists only of casing-related lints.

NON_SHORTHAND_FIELD_PATTERNS
WHILE_TRUE

Future compatibility++.
Errors that are reported as lints for some reasons unknown to me.
See the question in #45424 (comment) as well.

CONST_ERR // ?
UNKNOWN_CRATE_TYPES // Deny-by-default
NO_MANGLE_CONST_ITEMS // Deny-by-default
NO_MANGLE_GENERIC_ITEMS

Restrictions.
Something generally reasonable that can be prohibited if necessary.

BOX_POINTERS // Allow-by-default
UNSAFE_CODE // Allow-by-default
UNSTABLE_FEATURES // Allow-by-default
MISSING_DOCS // Allow-by-default
MISSING_COPY_IMPLEMENTATIONS // Allow-by-default
MISSING_DEBUG_IMPLEMENTATIONS // Allow-by-default

Pedantic.
Something not bad enough to always report/fix.

UNUSED_RESULTS // Allow-by-default
UNUSED_IMPORT_BRACES // Allow-by-default
UNUSED_QUALIFICATIONS // Allow-by-default
TRIVIAL_CASTS // Allow-by-default
TRIVIAL_NUMERIC_CASTS // Allow-by-default
VARIANT_SIZE_DIFFERENCES // Allow-by-default
UNIONS_WITH_DROP_FIELDS

Obvious mistakes.
Prevent foot shooting, some can become hard errors in principle.

OVERFLOWING_LITERALS
EXCEEDING_BITSHIFTS // Deny-by-default
UNCONDITIONAL_RECURSION
MUTABLE_TRANSMUTES // Deny-by-default
IMPROPER_CTYPES
PLUGIN_AS_LIBRARY
PRIVATE_NO_MANGLE_FNS
PRIVATE_NO_MANGLE_STATICS

General purpose lints.

WARNINGS
DEPRECATED

[oli-obk]

The lint store has a function register_removed

[oli-obk]

Just removing these lints will cause users who mention them in allow/warn/deny statements to produce warnings about unknown lints. Does rustc have a removed lint list like clippy does?

[petrochenkov]

Yes, there's a list of removed lints, I'll add these lints into it.

[oli-obk]

I would like fine grained grouping. I think we have an open issue in clippy. People often complain that they are overwhelmed by the style lints, when they want to get to the bug lints first

[bors]

☔ The latest upstream changes (presumably #45489) made this pull request unmergeable. Please resolve the merge conflicts.

[petrochenkov]

Rebased and updated with a fix for #45424 (comment)

[oli-obk]

Lgtm now. Do we have an issue to discuss the fine grained lint groups and not clear cut "unused"-lints? #45424 (comment) as an issue would be good I think

@bors r+

[bors]

@oli-obk: 🔑 Insufficient privileges: Not in reviewers

[petrochenkov]

@bors r=oli-obk

[bors]

📌 Commit 0b7a8b5 has been approved by oli-obk

[petrochenkov]

@oli-obk

Do we have an issue to discuss the fine grained lint groups and not clear cut "unused"-lints?

I've created #45615 for this.

[bors]

⌛ Testing commit 0b7a8b5b146c002177ade20be0a29a2d34cafed3 with merge 518ec2985dd8476aabaa3966564073bc7a393464...

[bors]

💔 Test failed - status-travis

[kennytm]

@bors retry #43283 (arm-android test_process_output_fail_to_start 30min no output)

[01:10:10] test process::tests::test_process_output_fail_to_start has been running for over 60 seconds


No output has been received in the last 30m0s, this potentially indicates a stalled build or something wrong with the build itself.
Check the details on how to adjust your build configuration on: https://docs.travis-ci.com/user/common-build-problems/#Build-times-out-because-no-output-was-received

The build has been terminated

[bors]

⌛ Testing commit 0b7a8b5b146c002177ade20be0a29a2d34cafed3 with merge 5b370c494c9f0039676c7b72c21206d960155e95...

[bors]

💔 Test failed - status-travis

[kennytm]

The use of :vis have broken clippy. But should the feature gate error be emitted when compiling clippy, as a user of the macro instead of the provider? cc #41022 @durka.

Details

[00:34:29]    Compiling clippy_lints v0.0.166 (file:///checkout/src/tools/clippy/clippy_lints)
[00:34:29] error: :vis fragment specifier is experimental and subject to change (see issue #41022)
[00:34:29]  --> <declare_lint macros>:1:3
[00:34:29]   |
[00:34:29] 1 | ( $ vis : vis $ NAME : ident , $ Level : ident , $ desc : expr ) => (
[00:34:29]   |   ^^^^^^^^^^^
[00:34:29]   |
[00:34:29]   = help: add #![feature(macro_vis_matcher)] to the crate attributes to enable
[00:34:29] 
[00:34:29] error: :vis fragment specifier is experimental and subject to change (see issue #41022)
[00:34:29]  --> <declare_lint macros>:1:3
[00:34:29]   |
[00:34:29] 1 | ( $ vis : vis $ NAME : ident , $ Level : ident , $ desc : expr ) => (
[00:34:29]   |   ^^^^^^^^^^^
[00:34:29]   |
[00:34:29]   = help: add #![feature(macro_vis_matcher)] to the crate attributes to enable
[00:34:29] 
[00:34:30] error: aborting due to previous error
[00:34:30] 
[00:34:30] error: Could not compile `clippy_lints`.
[00:34:30] warning: build failed, waiting for other jobs to finish...
[00:34:30] error: aborting due to previous error
[00:34:30] 
[00:34:30] error: Could not compile `clippy_lints`.
[00:34:30] warning: build failed, waiting for other jobs to finish...
[00:34:41] error: build failed

[petrochenkov]

@bors r=oli-obk

[bors]

📌 Commit bf0cdb5 has been approved by oli-obk

[durka]

Hmm. Ideally the writer of the macro should need the feature gate, not the user. But AFAIK pretty much all checking of macros is left until the last possible second. For example the current version of error_chain has an undefined capture and nobody noticed until yesterday (and only because they wrote a stricter compiler).

[durka]

Let's just stabilize :vis and remove the issue 😛

[bors]

⌛ Testing commit bf0cdb5 with merge dae6868...

[bors]

☀️ Test successful - status-appveyor, status-travis
Approved by: oli-obk
Pushing dae6868 to master...

[Manishearth]

Uh, please at least try to fix tools that break (even if you mark it as broken in the interim).

In this case for some reason #![feature(macro_vis_matcher)] does not fix the problem in clippy; it continues to give the same error despite being allowed already.

[Manishearth]

Oh, nvm, I made a mistake. Sorry.